Large Language Models (LLMs) increasingly serve diverse global audiences, making it critical for responsible AI deployment across cultures.While recent works have proposed various approaches to enhance cultural alignment in LLMs, a systematic analysis of their evaluation benchmarks remains needed.We propose a novel framework that conceptualizes alignment along three dimensions: Cultural Group (who to align with), Cultural Elements (what to align), and Awareness Scope (how to align: majority-focused vs. diversity-aware).Through this framework, we analyze 105 cultural alignment evaluation benchmarks, revealing significant imbalances: Region (37.9%) and Language (28.9%) dominate Cultural Group representation; Social and Political Relations (25.1%) and Speech and Language (20.9%) concentrate Cultural Elements coverage; and an overwhelming majority (97.1%) of datasets adopt majority-focused Awareness Scope approaches.In a case study examining AI safety evaluation across nine Asian countries (Section 5), we demonstrate how our framework reveals critical gaps between existing benchmarks and real-world cultural biases identified in the study, providing actionable guidance for developing more comprehensive evaluation resources tailored to specific deployment contexts.1 We extend the demographic proxies from Adilazuarda et al. (2024) to include appearance, sexual orientation, health, socioeconomic status, online communities, and time periods (e.g., dynasties) to better accommodate our dataset analysis.2 From Thompson et al. (2020), semantic domains include: food and drink, clothing and grooming, speech and language, modern world, physical world, the house, basic actions and technology, agriculture and vegetation, animals, social and political relations, emotions and values, kinship, and cognition, with names from Adilazuarda et al. (2024).We extend this list to include arts, mental perception, and current and historical events based on our dataset findings.3 These can be understood as prototypical representations of cultural knowledge (cf.Rosch and Mervis 1975; see also Shore 1996).We also recognize that some papers we classified as majority-focused do consider multiple annotator perspec-

Shayan Ali Akbar, Md Mosharaf Hossain, Tess Wood, Si-Chi Chin, Erica M Salinas, Victor Alvarez, Erwin Cornejo. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

In this paper, we describe a novel framework to recommend personalized intervention strategies to minimize 30-day readmission risk for heart failure (HF) patients, as they move through the provider's cardiac care protocol. We design principled solutions by learning the structure and parameters of a multi-layer hierarchical Bayesian network from underlying high-dimensional patient data. Next, we generate and summarize the rules leading to personalized interventions which can be applied to individual patients as they progress from admit to discharge. We present comprehensive experimental results as well as interesting case studies to demonstrate the effectiveness of our proposed framework using large real-world patient datasets on Microsoft Azure for Research platform.

Developing holistic predictive modeling solutions for risk prediction is extremely challenging in healthcare informatics. Risk prediction involves integration of clinical factors with socio-demographic factors, health conditions, disease parameters, hospital care quality parameters, and a variety of variables specific to each health care provider making the task increasingly complex. Unsurprisingly, many of such factors need to be extracted independently from different sources, and integrated back to improve the quality of predictive modeling. Such sources are typically voluminous, diverse, and vary significantly over the time. Therefore, distributed and parallel computing tools collectively termed big data have to be developed. In this work, we study big data driven solutions to predict the 30-day risk of readmission for congestive heart failure (CHF) incidents. First, we extract useful factors from National Inpatient Dataset (NIS) and augment it with our patient dataset from Multicare Health System (MHS). Then, we develop scalable data mining models to predict risk of readmission using the integrated dataset. We demonstrate the effectiveness and efficiency of the open-source predictive modeling framework we used, describe the results from various modeling algorithms we tested, and compare the performance against baseline non-distributed, non-parallel, non-integrated small data results previously published to demonstrate comparable accuracy over millions of records.

We present a system called Risk-O-Meter to predict and an- alyze clinical risk via data imputation, visualization, predic- tive modeling, and association rule exploration. Clinical risk calculators provide information about a person's chance of having a disease or encountering a clinical event. Such tools could be highly useful to educate patients to understand and monitor their health conditions. Unlike existing risk calcu- lators that are primarily designed for domain experts, Risk- O-Meter is useful to patients who are unfamiliar with medi- cal terminologies, or providers who have limited information about a patient. Risk-O-Meter is designed in a way such that it is flexible enough to accept limited or incomplete data in- puts, and still manages to predict the clinical risk efficiently and effectively. Current version of Risk-O-Meter evaluates 30-day risk of hospital readmission. However, the proposed system framework is applicable to general clinical risk pre- dictions. In this demonstration paper, we describe different components of Risk-O-Meter and the intelligent algorithms associated with each of these components to evaluate risk of readmission using incomplete patient data inputs.

A Knowledge Base (KB) stores, organizes, and shares information pertinent to entities (i.e. KB nodes) such as people, organizations, and events. A large KB system, such as Wikipedia, relies on human curators to create and maintain the content in the systems. However, it has become challenging for human curators to sift through the rapidly growing amount of information and filter out the information irrelevant to a KB node. The area of Knowledge Base Enhancement (KBE) aims to explore and identify automatic methods to assist human curators to accelerate the process. KBE can be viewed as a special case of Information Filtering (IF). However, the lack of high-quality labelled data introduces a major challenge to train a satisfying model for the task. Transfer learning provides solutions to the problem and has explored applications in the area of text mining, whereas direct application to KBE or IF remains absent.

Click logs from search engines provide a rich opportunity to acquire implicit feedback from users. Patterns derived from the time between a posted query and a click provide information on the ranking quality, reflecting the perceived relevance of a retrieved URL. This paper applies the Kaplan-Meier estimator to study click patterns. The visualization of click curves demonstrates the interaction between the relevance and the rank position of URLs. The observed results demonstrate the potential of using click curves to predict the quality of the top-ranked results.